tphakala/birdnet-go

Self-hosted realtime soundscape analyser for birds, bats and other wildlife. Multi-model local AI inference, runs 24/7 on a Raspberry Pi.

What it solves

BirdNET-Go is a self-hosted, 24/7 real-time soundscape analyzer designed to identify birds, wildlife, and bats. It eliminates the need for cloud-based analysis by providing local AI inference, allowing users to monitor their environment's biodiversity in real-time without sending audio data to external servers.

How it works

The system ingests audio from soundcards or network streams (RTSP/RTSPS) and processes it using a variety of AI models. It supports multiple models running in parallel—such as BirdNET v2.4, Google Perch v2, and BattyBirdNET—to increase detection confidence through cross-model agreement. The software includes a web-based dashboard for live spectrograms, detection heatmaps, and a configurable alert engine that sends notifications to platforms like Discord, Telegram, and Home Assistant via MQTT.

Who it’s for

It is designed for birdwatchers, wildlife researchers, and hobbyists who want to deploy local, automated audio monitoring stations, particularly on low-power hardware like the Raspberry Pi.

Highlights

  • Multi-model AI gallery: Install and switch between BirdNET, Google Perch, and bat classifiers without rebuilding the app.
  • Cross-model consensus: Boosts confidence by comparing detections across different AI models.
  • High-fidelity audio support: Supports sample rates up to 256 kHz for ultrasonic bat detection.
  • Comprehensive alert system: Routes detections to a wide array of services including Slack, ntfy, and webhooks.
  • Local-first design: Operates entirely locally by default with an embedded TFLite model.
  • Production-ready ops: Includes OIDC/SSO, TLS management, and a system health diagnostics page.

Related

  • Project
    woheller69/whoBIRD

    An Android app that uses machine learning to identify over 6,000 bird species by their sound in real time and entirely offline.

  • Project
    birdnet-team/BirdNET-Analyzer

    A deep learning tool for avian diversity monitoring that analyzes audio files to identify bird species across thousands of species.

  • Project
    microsoft/Biodiversity

    A collection of open-source AI tools for biodiversity monitoring, providing models for animal detection in camera-trap imagery, bioacoustics, aerial views, and sonar.

  • Dispatch
    Google DeepMind Perch Update: Advancing Bioacoustics for Endangered Species

    Google DeepMind has released an updated open-source Perch model that improves bird species prediction and expands bioacoustic analysis to mammals, amphibians, and underwater environments like coral reefs.

  • Dispatch
    Google DeepMind Mapping, Modeling, and Understanding Nature with AI

    Google DeepMind has announced new AI research and tools for predicting deforestation risk, mapping species distributions at scale, and identifying animal vocalizations via the Perch 2.0 bioacoustics model.